2be692687f
Fix color style properties lookup when the color resolution fails
...
The cached GValue ended up with a GtkSymbolicColor type, so
the contained value was bogus. Noticed by Mitch.
2011-02-07 19:56:17 +01:00
23f9bdd586
Fix problems with state propagation
...
Sensitivity changes were not properly propagated down the
hierarchy. There were two issues here:
a) correctly identifying when a state change request affects
sensitivity
b) not filtering out sensitivity in gtk_widget_propagate_state(),
since gtk_widget_set_sensitivity() uses that to do its work
https://bugzilla.gnome.org/show_bug.cgi?id=641431
2011-02-07 11:27:50 -05:00
1c20c93423
Optimize queue_resize_on_widget()
...
Optimized GtkSizeGroup code that is invoked for every queued resize
and every request that is not previously cached by trading qdata on
widgets for 3 extra bitfields on the GtkWidgetPrivate structure.
2011-02-08 00:39:51 +09:00
db97f8660a
Use faster new G_VALUE_COLLECT_INIT variant instead G_VALUE_COLLECT
...
Fixes https://bugzilla.gnome.org/show_bug.cgi?id=641621
2011-02-07 13:14:47 +00:00
beb8c290d1
Always chain up in ::style-updated
...
This was not handled consistently, but the default handler
does useful things, so we should always chain up.
2011-02-07 07:30:37 -05:00
14eb34874a
configure.ac: Remove some more obsolete macros
...
AC_TYPE_SIGNAL: obsolete when assuming C89 or better
AC_HEADER_SYS_WAIT: current systems are compatible with POSIX
2011-02-06 22:52:55 +00:00
8a965b9786
configure.ac: Use AC_COMPILE_IFELSE instead deprecated AC_TRY_COMPILE
2011-02-06 20:36:37 +00:00
c734bb8e83
configure.ac: Use AS_HELP_STRING instead deprecated AC_HELP_STRING
2011-02-06 20:36:37 +00:00
899019d56a
configure.ac: Remove some obsolescent macros
...
AC_C_CONST: current C compilers support const
AC_TYPE_SYGNAL: obsolete when assuming C89 or better
AC_HEADER_STDC: current systems have conforming header files
2011-02-06 20:36:37 +00:00
9bd175536b
Add a testcase for state propagation problems
2011-02-06 14:14:48 -05:00
24c37e966c
Updated Norwegian bokmål translation.
2011-02-06 19:14:53 +01:00
733731e80b
No more gtk_plug_*/gtk_socket_* symbols on win32
2011-02-06 15:07:14 +01:00
b41ee31a16
win32: no more GdkNativeWindow and related changes
...
- replace GdkNativeWindow with HWND, remove type casts
- no more GdkDisplayClass::get_drag_protocol but GdkWindowImpl::get_drag_protocol
- remove *_client_message*()
2011-02-06 15:07:13 +01:00
f55ccce2d9
win32: msvc build update
2011-02-06 15:07:12 +01:00
e39e8acd67
gtk: gtk_cell_renderer_get_state(): check the widget's effective sensitivity
2011-02-06 14:47:58 +01:00
1c0f85a813
gtk: gtk_cell_renderer_get_state(): a row can be both selected and insensitive
...
so pull the SELECTED state flag out of the !sensitive branch. Also,
don't make FOCUSED depend on SELECTED here, it's up to the widget to
decide whether or not that is possible.
2011-02-06 14:45:49 +01:00
1aaed01d4f
Updated Polish translation
2011-02-06 14:11:57 +01:00
330158977d
gtk: make the .trough background-color darker() as it was in GTK+ 2.x
2011-02-06 13:48:42 +01:00
42778894c0
s/state_changed/state_flags_changed/ in GtkComboBox
2011-02-06 11:21:42 +01:00
baa6343233
Handle insensitive cell renderer flags in gtk_cell_renderer_get_state()
2011-02-06 11:21:42 +01:00
217d3fd796
GtkStyleContext is meant to have a GdkScreen
...
It is used to get the default providers, without them
the style context can't do much. A check for NULL screen
is done before any sensitive call to
gtk_style_context_set_screen(), in the hope that any widget
will open the display before doing anything related to
styling. Fixes bug #641429 , reported by Bastien Nocera.
2011-02-06 11:21:42 +01:00
f23585b7bb
gtk: bail out in render_background_internal() if the extents are <= 0
...
otherwise we run into scaling the cairo_t into a borken state that's
not fixable by cairo_restore().
2011-02-06 11:17:59 +01:00
858ba35873
gtk: remove function gtk_entry_queue_draw() and call gtk_widget_queue_draw()
...
the old function was a leftover from when GtkEntry had a window
and was now invalidating the entire toplevel on each entry change.
2011-02-06 11:17:58 +01:00
c1bc521e92
gtk: get rid of two cases of invalidating the spinbutton when not needed
2011-02-06 11:17:58 +01:00
69e4a3f235
Updated Galician translations
2011-02-06 00:26:36 +01:00
bfaf472e70
Avoid tons of warnings from stylecontext initialization without screen
...
This reverts part of commit 4dc9b29
2011-02-05 11:21:21 -05:00
2e093cdc20
Demonstrate the problem with a workaround
2011-02-05 10:58:50 -05:00
c0ba1c96d4
Ensure gtk_render_background()/frame() yield the same shape
...
On rounded boxes the background and the frame wouldn't fit in
perfectly, leaving some visual noise in the corners. This is now
gone.
2011-02-05 15:26:40 +01:00
4c40349b92
Use cairo_push_group() to mix backgrounds during transitions
...
It also helps remove the cairo_clip() call to achieve the rounded
shape.
2011-02-05 15:26:40 +01:00
c990699922
gtk: don't style GtkLabel:selected's color separately
...
or labels in selected state will not have a different color. Reviewed
by Carlos.
2011-02-05 14:10:18 +01:00
ceeb50be9a
Don't leak a cell area context
2011-02-05 02:03:35 -05:00
ab66ac50cf
Migration guide additions
...
Some information about plugs and sockets, and event filters.
2011-02-05 01:56:19 -05:00
19092fe44f
Add a testcase for cell positions
2011-02-05 01:24:21 -05:00
8f870ddcdc
gtk-demo: Use symbolic icon that exists
...
For the Images test.
2011-02-05 03:20:27 +00:00
5f04bf6429
Simple test for display-less widget creation
2011-02-04 22:06:25 -05:00
4dc9b294d8
Don't assume a screen exists before realize()
...
Fix for https://bugzilla.gnome.org/show_bug.cgi?id=641429
2011-02-04 22:03:11 -05:00
2b2e607492
Improve GtkApplication migration guide
...
As pointed out in bug 641517, the examples did not actually
do the same.
2011-02-04 21:21:14 -05:00
3c1ff82b21
Fix a memory leak in GtkCSSProvider
...
Found by Felix Riemann
https://bugzilla.gnome.org/show_bug.cgi?id=641558
2011-02-04 21:13:56 -05:00
eaba2cd468
XI2: translate group state back into core state
...
This is necessary to make the key event translation work
as expected.
https://bugzilla.gnome.org/show_bug.cgi?id=641367
2011-02-04 21:11:40 -05:00
d4c1dbe3d7
Pass group state to _gdk_x11_device_xi2_translate_state
...
This will be necessary to make groups work with XI2.
2011-02-04 21:11:40 -05:00
2c352f51a0
Share code to fill GdkEventKey.string
...
This code was duplicated between the core and XI2 device manager
implementations.
2011-02-04 21:11:40 -05:00
4960a81082
Updated Arabic translation
2011-02-05 00:32:07 +02:00
3cb758a7b9
Add an empty English translation for properties
...
Otherwise the build is unhappy about en in LINGUAS
2011-02-04 17:18:24 -05:00
9a7708952c
Add an English translation
...
Currently, this just includes translations for keysyms.
2011-02-04 15:35:44 -05:00
b6783933c9
Strip XF86 prefix from untranslated keysyms
...
When we don't have translations, at least remove the
ugly XF86 prefix.
2011-02-04 15:34:36 -05:00
08f90b71a2
Add translation entries for more common keysyms
...
In particular, the XF86 keysyms can benefit from a translation
that strips the prefix.
2011-02-04 15:33:36 -05:00
9ba009a8eb
Remove debug spew
2011-02-04 14:05:22 -05:00
272aa7bc55
XI2: don't translate button releases as scroll events
...
This fixes scroll-by-2 sympoms seen in combo boxes.
https://bugzilla.gnome.org/show_bug.cgi?id=639846
2011-02-04 12:54:44 -05:00
03df034684
Remove latest bits of GtkItemFactory
2011-02-04 17:30:11 +00:00
012bc98936
Fixed gtk_recent_chooser_menu_set_current_filter to update the local pointer.
...
Before this patch, unsetting the filter manually before disposing
the recent chooser menu (or unsetting the filter twice), would cause
warnings (or invalid memory accesses).
2011-02-05 01:00:55 +09:00
